Feb 1, 2009 - depiction of the American school, IPE in the United States is increasingly pos- itivist, quantitative ... ISSN 0969-2290 print/ISSN 1466-4526 online C ..... very good job accounting for trade patterns, aid flows, persistent under-.
Review of International Political Economy 16:1 February 2009: 6–33
The American school of IPE Daniel Maliniak1 and Michael J. Tierney2 1
University of California–San Diego, San Diego, CA, USA 2 College of William & Mary, Williamsburg, VA, USA
ABSTRACT This paper uses the results of the Teaching, Research, and International Policy (TRIP) project: a multi-year study of the international relations (IR) field in order to discern the major characteristics of international political economy scholarship in the United States today. It finds that, like Benjamin Cohen’s depiction of the American school, IPE in the United States is increasingly positivist, quantitative, and liberal in orientation. It employs data from a journal article database that tracks trends in publication patterns. It also analyzes data from two surveys of IR scholars in the United States and Canada that were conducted in the fall of 2006.
KEYWORDS IPE; liberal theory; journal rankings; American school; epistemology; methodology.
In a keynote speech to the inaugural meeting of the International Political Economy Society (IPES) at Princeton University in November 2006, Benjamin Cohen argued that there were at least two distinct schools of thought that have adopted the moniker ‘IPE’ (international political economy) – the ‘British school’ and the ‘American’ school.1 According to Cohen, the intellectual evolution of the IPE field has produced an American school characterized by ‘the twin principles of positivism and empiricism’ and a British school driven by a more explicit normative, interpretive, and ‘ambitious’ agenda (Cohen, 2007: 198–200). These schools diverge in the ontologies, epistemologies and normative stances that each employs to study the same subject – ‘the complex linkages between economic and political activity at the level of international affairs’ (Cohen, 2007: 197). In Cohen’s view, IPE is increasingly fractured along conceptual and geographical lines, yet the American and British schools remain complementary. As such, he ends his reflection on IPE’s transatlantic divide by calling Review of International Political Economy C 2009 Taylor & Francis ISSN 0969-2290 print/ISSN 1466-4526 online http://www.tandf.co.uk DOI: 10.1080/09692290802524042
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E
for mutual respect, learning and a ‘meeting of the minds’ (Cohen, 2007: 216–17). Cohen’s speech, subsequent article in this journal (Cohen, 2007), and book (Cohen, 2008) have sparked a vibrant and contentious debate on the origins, character, and even desirability of a transatlantic IPE divide.2 Our reaction was less visceral than most. Nonetheless, we were provoked by Cohen’s depiction of the field and inspired to treat his characterizations of each school, based upon his interpretive intellectual history, as hypotheses that merit further testing. This interest coincided with our ongoing project on Teaching, Research, and International Policy (TRIP): a multi-year study of the international relations field in the United States and Canada. In the TRIP project, we employ data gathered from two extensive surveys of international relations (IR) scholars and a new journal article database that codes all the articles in the 12 leading political science journals that publish articles in the subfield of IR. Our journal article database covers the years between 1980 and 2006 (see below for note on methodology). While the TRIP project was not designed to provide the definitive test of Cohen’s thesis, it does provide us with some leverage on his claims. More importantly, since the TRIP project utilizes distinctive data collection, coding, and analysis methods – compared to Cohen’s methods – it provides a potentially powerful cross-check on Cohen’s findings about the nature of the IPE subfield and its purported divide. Thus, the purpose of this article is to use the TRIP data to investigate the American IPE school upon which much of Cohen’s argument is premised and from which the lively discussion surrounding the American versus British school has sprung. Ultimately, unlike Cohen, we do not seek to persuade others of the existence of stark differences between IPE scholarship in the United States and Europe. Indeed, our data are limited to IPE scholars in the United States and Canada and to the top journals in the field of IR (as determined by their Garand and Giles impact scores).3 Thus we consciously refrain from making assertions about the nature of the British IPE and the existence or nature of any transatlantic divide.4 Moreover, we remain agnostic about the prospects or desire for transatlantic bridge building within IPE (we leave this debate to others, including those who are contributing to this special issue of RIPE). Yet we are convinced that good bridges require solid foundations; and solid foundations require a clear understanding of the shores on which the foundations are built. Using the TRIP data, we can at least say something systematic about the American shore. We have two specific objectives in this article. The first is to ‘test’ specific hypotheses derived from Cohen’s argument. If Cohen’s depiction of the American IPE school is consistent with the results of our survey and patterns of journal article publications, then his broader thesis about the IPE subfield are further validated and the implications of his argument 7
R E V I E W O F I N T E R N AT I O N A L P O L I T I C A L E C O N O M Y
merit the animated debate that we have already witnessed. If Cohen’s depictions are not consistent with our findings, we should question his underlying assumptions and direct further research into the ‘myth’ of the divide – questioning why some perceive a divide that is non-existent or quite small. Our second objective is to use the TRIP data to assess prominent trends in American – and to a lesser extent Canadian – IPE. The TRIP project is well equipped to do this, insofar as it is quite broad in scope. The survey questions and variables coded in the journal article database capture the essential human and institutional ‘demography’ of the IR field, as well as the paradigmatic, theoretical, methodological, and epistemological orientation of that field. We can parse out variables most relevant to the American IPE subfield. In doing so, we reveal some remarkable and sometimes surprising findings that raise numerous questions directly relevant to understanding the state of IPE in the United States and its place within the broader IR discipline. In this paper, we take particular note of four trends in American IPE: its institutional and human demography, its ‘paradigmatic personality’, the growing methodological homogeneity, and the surprising absence of any ‘ideational turn’ which is so prominent in the other subfields of IR. We expect this paper to generate more questions than answers. While some of our data is formatted so that it directly speaks to extant hypotheses, much of what follows simply describes patterns of behavior, publication, or the aggregated opinion of IPE scholars in the United States and Canada. We were quite surprised by some of our findings and expect them to provoke a variety of explanations, reflections on the past, and consequences for the future of the discipline. B R I E F NOT E ON P R O J E C T M E T H O D O L O G Y In order to describe the American school of IPE we utilize the Teaching and Research in International Policy (TRIP) project’s databases.5 First, we employ results from two surveys: one of American IR scholars from 2004 and one of American and Canadian scholars surveyed in 2006 in order to describe the research practices of IPE scholars in those institutions.6 We also report United States IPE scholars’ views on the broader IR discipline and on some pressing foreign policy issues. In order to distinguish IPE scholars from the broader IR community, we often compare the responses of these two groups. Second, we use the TRIP journal article database, which covers the top 12 journals in political science that publish research on international relations. The time series spans 1980–2007 (Maliniak et al., 2007a).7 Since publication in these journals is not limited to American IPE, this data source – unlike the survey – can help to describe both the American and British schools of IPE.8 The article database reveals which of the top 8
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E
12 journals publish the most (and most cited) articles within the IPE subfield. This database also allows us to identify trends in the substantive focus of IPE research, the rise and fall of paradigms in the IPE literature, the methods employed most frequently, direct comparisons between the IPE literature and the broader IR literature, and whether IPE generates theory and methods that diffuse into the rest of the IR literature or vice versa. In addition to a description of the American IPE subfield, we employ the TRIP journal article database in order to provide some preliminary tests of Cohen’s comparative hypotheses. Are non-American IPE scholars publishing work that is substantially different from their American IPE cousins? Is American work more positivist, quantitative and formal, while British and European IPE is more non-positivist, normative, and qualitative? If these claims are true on average, how large are the differences between American and British styles of IPE and are these differences growing or shrinking? While Canada is not in Europe, some preliminary research suggests that it may be somewhere between United States and Britain in terms of the sensibilities of scholars located there and in terms of the research they produce. The 2006 TRIP Survey included IR and IPE scholars at Canadian universities and they appear to fit more comfortably in Cohen’s ‘British school’ than in the American one right next door. The journals in the TRIP database are dominated by scholars at American institutions. One bit of evidence suggesting that there are two distinct IPE communities is the pattern of publication displayed in Figure 1. When we compare the percent of authors at non-US institutions in our sample to those publishing in RIPE, we see a dramatic difference. For every year over the past decade we observe less than 20% of non-US authors publishing IPE articles in the top 12 journals. This differs dramatically from the distribution of articles at RIPE, where we never observe less that 60% of non-US authors. Over the past ten years the percentage of US-based authors publishing in RIPE has dropped from around 40% to just 30%. So, if we accept the Murphy and Nelson characterization of RIPE as the flagship journal of British style IPE, then we have some evidence for a large and growing gap between British IPE and IPE published in the other leading journals.
T HE F I E L D : W HAT D O E S A M E R I C A N I P E L O O K L I K E ? The demography of American IPE How do IPE scholars differ from the rest of the field of international relations within the United States? Using answers to the TRIP surveys allows us to measure specific characteristics of the individuals who make up the IPE subfield. In some respects scholars who claim IPE as their primary or secondary issue area differ from the broader population of IR scholars, but 9
R E V I E W O F I N T E R N AT I O N A L P O L I T I C A L E C O N O M Y 100%
90%
80%
70%
60%
American authors in top 12 Non-American authors in top 12 American authors in RIPE Non-American authors in RIPE
50%
40%
30%
20%
10%
06 20
00 20
19
95
90 19
85 19
19
80
0%
Figure 1 Percent of US/Non-US Authors Publishing in Top 12 Journals versus RIPE. Note: For ease of presentation, all time series data in this paper are reported using three year rolling averages.
in other ways they are indistinguishable. IPE scholars are trained at different schools, they use different methods, they study different regions of the world, and they come from different regions of the world (specifically, they are far more international than their other IR colleagues at United States institutions). However, in other respects where we might expect variation across areas of study, we see very little. The percent of men and women studying IPE as IR is basically the same; IPE scholars are the same age on average as their IR counterparts, and they rank journals, PhD programs, and threats to United States national security about the same as the broader IR community. Overall, 30% of IR scholars in the United States do work in IPE.9 Specific schools have reputations for being particularly strong in IPE (Harvard, Berkeley, Princeton, UCSD, and UCLA often get mentioned at the APSA bar), but the conventional names today are not always the same programs that have produced the largest number of IPE scholars in the United States over the past 40 years. At minimum, this variation suggests that comparative advantages within the top PhD programs change over time. Although only 2% of all IR scholars received their doctoral training from Yale University, more IPE scholars (5%) trained at Yale than any other 10
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E
Table 1 Departments training the most IPE scholars Rank 1 2 2 4 4 4 4 8 9 9 9 9
University
Percent
Yale University Columbia University Cornell University Harvard University University of California, Berkeley University of California, Los Angeles University of Wisconsin University of Michigan MIT Princeton University Stanford University UNC Chapel Hill
5 4 4 4 4 4 4 3 3 3 3 3
program. In addition to Yale, several other institutions have produced proportionately more IPE scholars than IR scholars studying in other issue areas. For example, University of Wisconsin at Madison ranks fourth for IPE but only 13th overall, Princeton University is tied for tenth with University of North Carolina, but these two schools rank 15th and 26th in terms of the total number of IR scholars produced. In addition to what universities are training the next generation of graduate students, the article database reveals which programs produce the most IPE articles in the top 12 journals. We code the home department of authors upon publication of their article, and find that Harvard tops the list, with its scholars having penned 5% of all the IPE articles in the top journals since 1980. The top three schools for IPE are the same programs in order as IR generally. Strikingly, University of Colorado is tied for third in IPE articles, yet is 11th for the broader IR category.10
Table 2 Departments training the most IR scholars Rank 1 2 3 4 5 6 7 8 9 10
University
Percent
Columbia University Harvard University University of Michigan University of California, Berkeley Cornell University University of Virginia Ohio State University Stanford University MIT University of Chicago
5 5 4 3 3 3 3 3 3 2
11
R E V I E W O F I N T E R N AT I O N A L P O L I T I C A L E C O N O M Y
Table 3 Number of IPE articles produced since 1980 Rank 1 2 3 3 4 5 5 5 8 9 9
University
Percent
Harvard University Columbia University Stanford University University of Colorado University of Chicago Princeton University University of California, Los Angeles Yale University Duke University New York University Ohio State University
5 3 3 3 3 2 2 2 2 2 2
IPE scholars at United States institutions are neither younger, nor more diverse in terms of their gender than other IR scholars. On average, IPE scholars received their PhD two years later (1992) than the broader group of IR scholars. This is somewhat surprising, since both groups have the same average age, which implies that IPE scholars either start graduate programs at a later age or they take longer to obtain their degree than other IR scholars. However, the late start or extended stay in graduate school may pay off later as IPE scholars are more likely to hold the position of full professor (37%) than those studying in other subfields (33%). Similarly, we find no evidence of a gender distinction within IPE that is different than the general IR population.11 While the percentage of women in IR as a whole is 23%, the percent who study IPE is only 22%. Research on publication rates in political science and IR demonstrate that women publish less than their male colleagues and IPE provides no exception
Table 4 Number of IR articles produced since 1980 Rank 1 2 3 4 5 6 7 8 9 10
University
Percent
Harvard University Columbia University Stanford University Ohio State University University of Michigan Princeton University Yale University University of Illinois University of Chicago University of California, Los Angeles
6 3 3 2 2 2 2 2 2 2
12
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E
to this trend. Since the year 2000 only 14% of all authors of IPE articles published in the leading journals were women. Despite this fact, there is strong evidence from the TRIP survey that IPE scholars value the research of women to a greater extent than other IR scholars do. More women appear in the various top 25 lists for greatest impact on the field (3), most interesting work (6) and most influential on your own research (4). In all three of these categories IPE scholars are more likely to list women than are IR scholars who study other issue areas.12 Where is IPE research published? Within the IR literature, articles with an IPE focus make up only 13% of those published since 1980 (despite 30% of IR scholars in the United States reporting their first or second field as IPE). Over this period, IPE’s share of articles in the top 12 journals has ranged from a high of 20% in 1984 and 1985, to a low of 5% in 1994. Within the journals analyzed there is significant variation. Thirty-seven percent of IPE articles since 1980 are found in the pages of International Organization (IO), with International Studies Quarterly (ISQ) and World Politics containing 22% and 10%, respectively. Perhaps not surprisingly then, IPE scholars rank IO and ISQ as two of the journals they read most often in their area of expertise, with 53% and 34%, respectively, and World Politics sixth, with 15%.13 American Political Science Review (APSR) and International Security (IS) both rank in the top five despite accounting for only 4% and 1% of IPE literature, respectively. For the APSR, this is somewhat unsurprising given the limited space devoted to IR in general. Within APSR, IPE articles account for 15% of published IR articles. The same cannot be said for IS, whose sole mission is publishing IR articles with a security focus. Over a 26-year period, less than 1% of the articles published by IS fit into the IPE category. In contrast, 18% of IO’s articles deal with issues of international security. In addition to looking at the number and percentage of IPE articles published in various journals, we also used citation counts in order to determine which journals publish IPE articles that have the greatest impact on the way other scholars think about their work. Despite IS’s exclusion of IPE-focused articles, since 1990, they hold the highest average number of cites per IPE article, 57.14 However, this is a result of the one IPE article that International Security published in the sample we coded, Michael Mastanduno’s ‘Do Relative Gains Matter? America’s Response to Japanese Industrial Policy’. For the other journals, IPE articles are cited less frequently (on average) than other IR articles in the same journals. The lone exception is the other security-focused journal, Security Studies, whose four IPE articles are cited an average of once more than all other articles in the journal. Not surprisingly, the three journals that publish the most IPE are also the three journals with largest number of aggregate citations of their 13
R E V I E W O F I N T E R N AT I O N A L P O L I T I C A L E C O N O M Y
IPE work. International Organization accounts for 49% of all the citations of IPE articles over the past 17 years, World Politics accounts for 15% and ISQ for 10%. The main outlier here seems to be ISQ, which publishes 22% of the IPE literature, yet only accounts for 10% of the citations. This may be a good proximate measure for the overall impact of the IPE work being published in a given venue. Therefore, the third best outlet of highly cited IPE work is APSR, which publishes 4%of the IPE literature but accounts for 9% of the total citations of IPE articles. Paradigmatic personality of American IPE In 1991, Susan Strange argued that the study of IPE should be ‘an open range, like the old Wild West, accessible . . . to literate people of all walks of life, from all professions, and all political proclivities’ (Strange, 1995: 33). We are pretty sure that Strange would not be happy with the state of the field today. In stark contrast to the broader field of IR, work in the area of IPE often falls within one of the four major paradigms identified in the TRIP article database (realism, liberalism, Marxism, or constructivism).15 In fact, only 45% of the broader IR literature comfortably fits into any of the major paradigms, while 73% of IPE articles are either realist, liberal, Marxist, or constructivist. This is not terribly surprising as many scholars have traditionally divided theories of IPE into three of these categories (Frieden and Lake, 1999; Gilpin, 1987; Katzenstein et al., 1998; Lake, 2006). IPE is, overall, more oriented around paradigmatic debates than the general field of IR. The paradigms preoccupying IPE work have changed quite dramatically over time. Historically, Marxism played an important role in the inception of modern IPE. However, today it is clear that arguments rooted in the Marxist paradigm almost never appear in the top 12 journals.16 Although the timeframe of our study does not cover the pre-1980 era, evidence of Marxism’s previous importance is visible at the beginning of our time series. As seen in Figure 2, articles in the Marxist tradition represent the largest proportion of the IPE literature in the early 1980s, but fall from their height of 39% in 1980 to 11% by 1986. This decline pre-dates the fall of the Soviet Union, suggesting that epistemic factors (rather than political ones) were driving the decline. David Lake (2006) suggests, Marxist political economy models simply were not doing a very good job accounting for trade patterns, aid flows, persistent underdevelopment in Africa, or the other substantive issues that IPE scholarship attempts to address. Moreover, according to Lake, Marxist approaches ‘failed to develop a unified, logically consistent, and empirically robust theory of under-development, or at least one that could compete in rigor and explanatory power with neoclassical economic theories of growth and development’ (Lake, 2006: 760). Since 1990 Marxist IPE has never 14
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E 80%
Realist Liberal Marxist Constructivist Non-Paradigmatic Atheoretic/None
70%
60%
50%
40%
30%
20%
10%
06 20
00 20
95 19
90 19
85 19
19
80
0%
Figure 2 Paradigms employed within IPE, 1980–2006.
represented more than 10% of the overall IPE literature and in recent years we observe no Marxist IPE articles being published in any of the leading journals. So what has replaced Marxism as the paradigm of choice? With the exception of period from 1980 to 1983, liberal research has been the most utilized paradigm by scholars publishing work on IPE. From 1988 to 1994 and 1996 to the present the majority of articles published in IPE employed a theory that fit within the liberal paradigm. This rise and dominance of liberal work comes at the expense of the atheoretic and Marxist literature, both of which have seen significant declines over the period we analyze. The fall of atheoretic work in IPE closely mirrors that of the IR field in general (Maliniak et al., 2007a); however, IPE has always been the most theoretically oriented of all the subfields. While we observe declines in atheoretic work across the board, there have been no atheoretic IPE articles published in the leading journals over the past nine years! This is not true for security, environment, human rights, international organization, international law, or other subfields within IR. The other two major paradigms, constructivism and realism, are dwarfed in comparison to liberalism and do not reach over 11% in any given year since 1980, and not over 6% in any of the past ten years. The overall picture of IPE work contrasts starkly to the IR field more broadly, which is made 15
R E V I E W O F I N T E R N AT I O N A L P O L I T I C A L E C O N O M Y
up of work primarily outside the four major paradigms (Maliniak et al., 2007a). While two of the traditional paradigms of IPE – liberalism and Marxism – individually hold a plurality of articles published for a portion of the time series, realism is represented in very little IPE research. Realist IPE is largely non-existent through the 1980s and peaks in 1992 and 1996 at just 8%.17 This is somewhat ironic since the standard IPE textbooks used in graduate seminars (Frieden and Lake, 1999; Gilpin, 1987) present realism as if it were one of the three main paradigms in the study of IPE. One explanation for this situation may be that although few Realist IPE scholars exist, the ideas of certain realists, such as Robert Gilpin, Stephen Krasner, and Joseph Grieco, have been used as the counter arguments, framing devices, and straw-men for many liberal and Marxist accounts. Below we provide some indirect evidence in support of this interpretation. A different way of getting at the impact of various paradigmatic ideas is to survey scholars about the individuals and the journals that have the biggest impact on their thinking. Among those scholars listed as having the most profound impact on self-identified IPE scholars, the top three reflect a great deal of theoretical diversity: Robert Keohane (19%), Robert Gilpin (12%) and Alexander Wendt (8%) all represent distinct paradigmatic commitments. Further inspection of Table 5 illustrates the diversity of individuals ‘having the most profound impact’ on IPE scholars (Maliniak et al., 2007c). Citation data provide further evidence for the claim that realism in IPE represents more of a straw-man than a research program. Realist IPE articles on average are cited more times than any other category – paradigmatic, non-paradigmatic or atheoretic articles. This finding is consistent with the hypothesis that many authors frame counterarguments in a traditional paradigmatic manner. Having a relative dearth of realist IPE articles from which to choose, authors find it necessary to continue trashing the same realist articles. In addition, the realist IPE literature is more frequently cited (20.5) than the rest of the realist literature (17.3) in IR by roughly three cites per article. Non-paradigmatic IPE articles also are more cited in IPE than in the IR literature generally, but that is only by just under one cite per article (0.8). Being true for both Marxist and liberals (2.5 and 5.7 fewer cites per IPE article, respectively), this suggests that paradigmatic work outside of an issue area in which it is traditionally strong may be a key to receiving more citations; however, they may all be negative citations that take the form, ‘Can you believe how dumb professor Smith is?’ Moreover, the data on the variable ‘alternative paradigms taken seriously’18 in an article suggest that citations are more likely cursory mentions and not the result of authors engaging alternative paradigmatic positions. With this strict definition, we do not just code those articles that include a citation of work from another paradigm; rather the author must take 16
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E
Table 5 Most Profound impact on your own research Rank 1 2 3 4 4 6 7 7 7 10 11 11 11 14 14 14 14 18 18 18 21 22 23 24 25
Scholar
Percent
Robert Keohane Robert Gilpin Alexander Wendt Joseph Nye Peter Katzenstein Robert Cox Jeffrey Frieden John Ruggie Kenneth Waltz Stephen Krasner David Lake G. John Ikenberry James Fearon Bruce Bueno De Mesquita Ernst Haas Samuel Huntington Susan Strange Helen Milner Immanuel Wallerstein Ronald Rogowski Andrew Moravcsik Beth Simmons Hans Morgenthau Hayward Alker Martha Finnemore
19 12 8 7 7 6 6 6 6 5 5 5 5 4 4 4 4 4 4 4 3 3 3 3 3
Note: The question said: ‘List up to four scholars who have had the most profound impact on your own research and the way that you think about IR.’ These are responses of scholars in the United States who indicated IPE as their primary or secondary field. For the Canadian list see Lipson et al. 2007.
theoretical implications of the other paradigm seriously, not just using it as a straw-man. It turns out that realism is taken seriously less often in IPE articles (13%) than in non-IPE articles (17%).19 Overall, just as in the general IR literature, a majority of IPE articles, 64%, do not take seriously any paradigm other than the paradigm they are advancing. The ‘mismatch’ between how the field is traditionally divided in text books and graduate syllabi versus the characteristics of published articles carries over into the description of IPE scholars as well. Of the 269 scholars who responded that their primary or secondary field was IPE in the 2006 survey, realists and constructivists accounted for 17% and 13%, respectively. This is greatly out of proportion to the number of articles published since 2000, in which only 3% are realist and 3% are constructivist. Although only 37% of IPE scholars described their paradigm as liberal or neo-liberal, liberal articles account for 66% of the IPE literature since 2000. 17
R E V I E W O F I N T E R N AT I O N A L P O L I T I C A L E C O N O M Y
Table 6 Greatest impact on the field of IR Rank 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Scholar
Percent
Robert Keohane Kenneth Waltz Alexander Wendt Joseph Nye Robert Gilpin John Mearsheimer James Fearon John Ruggie Samuel Huntington Stephen Krasner Bruce Bueno de Mesquita Robert Jervis Peter Katzenstein Robert Cox Helen Milner Jeffry Frieden Bruce Russett Hans Morgenthau Michael Doyle Susan Strange Immanuel Wallerstein J. Ann Tickner James Rosenau Mancur Olson Peter Gourevitch
64 41 28 16 16 12 11 11 11 11 10 8 7 6 5 5 4 4 4 4 3 3 3 3 3
Note: The question said: ‘List up to four scholars who have had the greatest impact on the field of international relations over the past 20 years.’ These are the responses of IPE scholars about the broader field.
This finding begs the question: Why is there such a preponderance of the liberal paradigm in IPE journal articles? Six percent of scholars in the 2004 and 2006 surveys describe their work as falling under the Marxist paradigm, while less than 1% of the IPE work published in the top 12 journals is Marxist since 2000. With very few scholars and a limited body of work, we are left to wonder why the field’s theoretical debates continue to be framed such that Marxism is still advanced as one of the three paradigmatic pillars upon which IPE stands. Methodology in ‘American style’ IPE For most political scientists in the United States, political economy is not conceived as a field, but rather as ‘the methodology of economics applied to the analysis of political behavior and institutions’ (Weingast and Wittman, 2006: 3). While David Lake defines IPE as ‘a substantive topic of enquiry, 18
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E 100% 90% 80%
Analytic/non-formal Counterfacutal Descriptive Experimental Formal Policy analysis Qualitative Quantitative
70% 60% 50% 40% 30% 20% 10%
00
06 20
19
20
95
90 19
85 19
19
80
0%
Figure 3 Methods employed in IPE articles, 1980–2006.
rather than a methodology in which economic models are applied to political phenomena’, IPE’s association with economics in the United States make it an obvious place to bring the tools of economics to bear on issues of political behavior and institutions. However, Lake also notes, ‘Where IPE was originally constrained to case studies or, at best, statistical tests based on very limited data, some of the best new research employs very large time series data-sets and powerful econometric tools’ (Lake, 2006: 772). If Lake is correct, we should see a substantial increase in quantitative work being published in IPE compared to past patterns. In fact, as illustrated in Figure 3, this is precisely the pattern we observe and the increase in quantitative work is dramatic. As illustrated in Figure 3, for most of the 1980s and 1990s qualitative methods were the most frequently used empirical tools in the IPE journal literature. A cursory glance at the data suggests that the methods used within IPE are similar to those used in the field overall, with qualitative as the mildly dominant method for most of the time period and quantitative methods ascending and taking the top spot more recently. However, the overall amount of quantitative work and its increasing use in recent years is much more pronounced within IPE than in the rest of IR (compare Figure 3 to Figure 4). This is all the more surprising if one considers the historical positions of qualitative versus quantitative work in the IPE subfield versus 19
R E V I E W O F I N T E R N AT I O N A L P O L I T I C A L E C O N O M Y 60%
50%
Analytic/non-formal Descriptive Experimental Formal Policy analysis Qualitative Quantitative
40%
30%
20%
10%
06 20
00 20
19
95
90 19
85 19
19
80
0%
Figure 4 Methods empoloyed in IR articles, 1980–2006.
all of IR. In the 1980s and early 1990s, qualitative methods were much more prevalent in IPE articles than in IR in general. Yet over the past five years, this relationship has reversed itself at a remarkable rate. By 2006, fully 90% of all IPE articles in our sample employed quantitative methods. Why has there been such a precipitous decline in qualitative IPE research in favor of quantitative work in the leading journals? While conventional wisdom suggests an increase in quantitative research, the rate of increase in quantitative work merits a more detailed discussion. If we look at the entire population of IR articles in the literature (see Figure 4), qualitative methods are never used in more than 43% of the articles in any given year, and only represent an average of 35% over the entire time period. Within the IPE literature, qualitative research peaks twice at 71% of the articles published in a given year and over the entire time period is used in an average of 48% of IPE articles. The real story here is that IPE literature employs both qualitative and quantitative methods at higher rates and at earlier points in the time series than the rest of the IR field. Apparently, IPE is a disciplinary leader in terms of the methods used – increases in these methods throughout the IR field lag about ten years behind their widespread adoption within IPE. When Jeffrey Frieden and Lisa Martin (2002) wrote that IPE had approached a ‘consensus on methods’, they were referring to methodology 20
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E
Table 7 Survey results on primary and secondary methods for IPE versus IR Primary method
Secondary method
Total
Method
IPE
Non-IPE
IPE
Non-IPE
IPE
Non-IPE
Quantitative Qualitative Formal modeling Experimental Counterfactual analysis Pure theory Legal/ethical analysis
27% 65% 2% 1% 1% 2% 2%
19% 70% 2% 1% 1% 2% 4%
31% 25% 15% 3% 25% 20% 19%
29% 22% 11% 8% 24% 17% 21%
58% 90% 17% 5% 26% 22% 20%
48% 92% 12% 9% 25% 19% 25%
in a broader sense – consensus on substantive questions, the relevance of institutions, and the utility of rationalist approaches to empirical questions. However, at this same time the field was approaching a consensus on methodology in narrower terms as well. Quantitative methods are now (by far) the most utilized method in IPE articles. The increase in quantitative work – while evident in all of IR – is particularly strong in IPE. In the ten years from 1993 to 2003, quantitative methods are used in 49% more articles (a jump from 21% to 70%). This increase comes at the expense of qualitative methods, which over the same period fell from 71% to 30%, on its way to currently being employed in only 11% of IPE articles in the 12 journals we coded. While the use of quantitative methods is not exclusively tied to economics, the prevalence of econometric techniques in the IPE literature suggests that Lake (2006) was correct in claiming that the methods of economics are the standard tool of IPE scholars publishing in the top journals. Further evidence of economic methods breaking into IPE can be seen in the increased use of formal models. From 1992, when there were virtually no formal methods used in IPE research, their prevalence has increased sharply to 22% of the IPE literature in 1998, and since has consistently hovered around 20%. As with quantitative work, this is substantially higher than the amount of formal work being done in other subfields of IR. In 2006 formal was the second most highly used method at 22%, 11% higher than qualitative work. Formal methods do not enjoy the same popularity in the IR literature as whole, and declined from 17% in 1995 to 12% in 2006. Given the methods employed in the IPE literature, it is unsurprising that a larger percentage of IPE scholars report using quantitative and formal methods in their research than other IR scholars do. The years of qualitative prominence prior to the rather sudden methodological shift in IPE are also reflected in the 90% of IPE scholars who continue to consider qualitative as their primary (65%) or secondary (25%) method. However, 21
R E V I E W O F I N T E R N AT I O N A L P O L I T I C A L E C O N O M Y
this seems to be the result of a generational gap. For those who received (or will receive) their doctorate in 2000 or later, 36% consider quantitative methods as their primary approach. IPE scholars who received their PhD in the 1990s rely less on quantitative methods (27%), and those who received their PhDs in the 1980s or earlier utilize quantitative methods even less (18%). This trend holds for the rest of IR scholars as well, but the differences are much starker within IPE. Twenty-three percent of nonIPE scholars who received (or will receive) their PhD after 1999 consider quantitative methods as their primary tool, with those who received their PhDs in the 1990s (20%) and those who received theirs in the 1980s or earlier (15%) each having progressively less reliance on quantitative methods. IPE also seems to exhibit two distinct periods of theoretically focused work. These ‘theory bumps’ of analytic/non-formal work (see Figure 3) feature articles devoted to analytic or theoretical issues without reference to significant empirical evidence or a formal model.20 These surges in theoretical articles occur once before the end of the Cold War, peaking in 1987 at 24%, only to fall away and represent little to none of the IPE literature from 1991 to 1994.21 The purely theoretic work again rises to prominence in 1996 at 16%, and steadily declines from that point to the present day. These theoretic works of the mid-1990s may be the result of the oft-lamented ‘paradigm wars’ in the IR literature. Regardless of their source, the following period sees a dramatic increase in quantitative, qualitative and formal IPE. One interpretation of these trends suggests that many of these theoretical debates provided fertile ground for empirical investigation, especially for quantitative researchers. Data on the citations of analytic/non-formal articles are suggestive, as the average number of citations is 30 per article, compared to formal, qualitative and quantitative research, which average eight, 15 and nine citations per article. The big ideas in analytic/non-formal articles may shape the research agendas in the subsequent periods that look more like Kuhnian ‘normal science’. What regions of the world do IPE scholars study? In this issue, Robert Keohane laments the fact that contemporary IPE scholars are not asking the ‘big questions’ about issues that are ‘important in the real world’ and are instead too focused on creating/testing extant theories or finding questions that could be addressed with sophisticated methodologies. One of the four ‘big questions’ that Keohane discusses is the rise of China. While we do not have a direct measure of the number of IPE scholars working on this question, we do have evidence on what IR and IPE scholars believe are the big questions and we have evidence on whether IPE scholars are publishing more on East Asia than their IR colleagues are. 22
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E
The TRIP survey indicates that IPE scholars in the United States are more likely to believe that East Asia is strategically important today compared to non-IPE people, 23% to 17%, while 6% fewer IPE scholars believe that the Middle East is the most strategically important region today. However, the two groups converge on the choice of which region will be strategically important in 20 years, 67% for IPE and 65% for non-IPE agree it will be East Asia. IPE scholars are only slightly less likely to think that the rise of China is one of the three most important foreign policy issues that will face the United States in the next ten years, 40% compared to 41% of all IR scholars. The regional focus of IPE is, perhaps not surprisingly, tied to economically developed countries. Thirty-five percent of IPE articles contain data or cases drawn from the United States, and, with obviously some overlap, 35% consider data and cases from Canada and Western Europe. The third largest region gaining attention in IPE is East Asia with 29%, followed closely by Global work, or those papers use data and cases covering every country or region in the world.
Ideational turn in IR . . . but not IPE Much has been made of the ‘ideational turn’ in IR over the past few decades, with many now considering constructivism to have overcome its subaltern
IPE only All IR
60%
50%
40% y = 0.005x + 0.3492 R2 = 0.7138
30%
20%
10%
Figure 5 The Material Turn in IPE.
23
20 06
20 04
20 02
20 00
19 98
19 96
19 94
19 92
19 90
19 88
19 86
19 84
19 82
19 80
0%
R E V I E W O F I N T E R N AT I O N A L P O L I T I C A L E C O N O M Y
Table 8 Types of ideational variables used in IPE vs IR
Religion Societal/political culture Organizational culture Beliefs Perceptions Identity International norms Domestic norms Other
IPE Scholars
Non-IPE Scholars
Difference
25% 66% 30% 50% 48% 46% 55% 36% 10%
30% 56% 28% 62% 57% 52% 50% 32% 7%
−5% 10% 2% −12% −10% −6% 5% 4% 2%
status and achieved paradigmatic popularity on par with realism or liberalism (Checkel, 1998). As illustrated by Figure 5, however, this turn appears to be almost nowhere evident in IPE articles published in the top journals, and, if evident at all, seems to be turning in the opposite direction.22 While at least 34% of the journal articles we coded contained ideational factors for every year since 1980, in the IPE articles this number never reaches above 25%, and for the majority of our time period it is below 20%. Even beyond the absolute differences between the two groups, the trends they display are markedly different. The overall IR literature displays a near linear increase (r 2 = 0.7138) over the time period, while the trend in IPE appears parabolic, falling off after 1995. In 2004, we asked IR scholars whether they emphasized the role of ideational factors in their research, and if so, which factors they emphasized. Overall, IPE scholars were 19% less likely than scholars in other subfields to report a focus on ideational factors in their research. However, the majority, 64%, do emphasize ideational factors, compared to the 83% of non-IPE scholars who employ ideational factors in their analyses. Not only do they employ ideational factors at different rates, but IPE and non-IPE scholars focus on different ideational variables in their work as well. As seen in Table 8, IPE scholars who do study ideational variables are more likely to focus on societal/political culture, international norms, domestic norms, and organizational culture, by ten, five, four and 2%, respectively.23 In our article database, we only code for one of these variables specifically: international norms. Analysis of this variable, as seen in Figure 6, shows that there is in fact an increase in the number of articles including discussion of international norms in IPE. However, while the increase is evident in absolute terms, IPE’s share of the literature including international norms has remained fairly stable over the past 27 years. In short, and with obvious influential exceptions, IPE scholarship has done little to contribute to the ‘ideational turn’ in IR research. 24
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E 100%
4
3 IPE Non-IPE Rolling Average
60%
2
40%
1 20%
0%
# of IPE articles including international norms
80%
0 1980
1990
2000
2006
Figure 6 Articles including International Norms, IPE and Non-IPE compared.
While IPE scholars differ from their IR colleagues on these two questions, they are remarkably similar in their response to whether or not their attention to ideational factors has changed over time; 53% of IPE scholars say that their attention has increased as compared to 51% of non-IPE scholars. Nevertheless, this increased attention does not seem to make its way into the articles in our database. Perhaps not surprisingly given these results on ideational questions, IPE also exhibits an epistemological trend that is much more positivist than the rest of the IR literature (although both are becoming more positivist over time). What is quite apparent on this front is the relatively drastic and absolute change in IPE epistemology, consistent with the quantitative methodology trends we have already discussed. As shown in Figure 7, IPE went from a low of 67% positivist articles in 1983 to 98% just three years later. This drastic change is far from an outlier, as positivist articles make up no less that 91% for the rest of the time period. IPE: an American enterprise? Many prominent scholars argue that IR is a ‘not so international discipline’,24 and is instead dominated by United States scholarship (Hoffman, 1977; Waever, 1998). Similar claims about IPE have focused on both American hegemony, but also on bipolarity, namely the split between the 25
R E V I E W O F I N T E R N AT I O N A L P O L I T I C A L E C O N O M Y 100% 90% 80% 70% IPE Positivist 60%
IPE Non/Postpositivist All IR positivist All IR non-positivist
50% 40% 30% 20% 10%
00
06 20
19
20
95
90 19
85 19
19
80
0%
Figure 7 Epistemology in IPE and IR compared.
American school – based on or around the journal IO – and the British school – known also as critical IPE and centered around RIPE and New Political Economy (Murphy and Nelson, 2001: 394). Among the differences between these two poles, as claimed by Murphy and Nelson (2001), are methodological and epistemological pluralities on the British side contrasted with a narrower view of science and obsession with broader theoretical debates on the American side – claims that mirror the comments of Cohen (2007). Admittedly, we do not have data on British IPE. However, we can help to confirm or refute claims about the American field with more specificity. Murphy and Nelson (2001, 397) argue ‘the IO school is more concerned with set debates about ways to study international relations’, albeit using little more than anecdotal evidence and a restatement of terminology coined in previous issues of OI. The amount of literature that fits within one of the four major paradigms (73%) helps to confirm this claim. In addition, Murphy and Nelson claim that neither the American nor the British schools adheres to the norms of science. The two norms they identify as characterizing a ‘scientific’ approach are the systematic collection and analysis of data and systematic theory building – or, essentially quantitative, empirical work and formal theory. They claim that political science and the sub-field of IR have indeed adopted a systematic empirical orientation, and that a cursory glance at any APSR is proof of this claim. But IPE is a 26
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E
Laggard, according to Murphy and Nelson. This claim is not true today, if it ever was. Our data clearly demonstrate that American IPE is now more quantitative and more formalized than any other area of IR. Moreover, this happened while retaining a commitment to theoretical traditions despite the claim that this was a major barrier (Murphy and Nelson, 2001: 403). CONCLUSION Cohen (2007, 2008) describes the field of IPE as one that is extremely divided in terms of theory, method, normative commitments, and epistemology. He believes that both sides of the transatlantic divide have a great deal to learn from each other and at various points he advocates dialogue, bridge building, and synthesis between these two schools of thought. Both Waever (1999) and Murphy and Nelson (2001) perceive a similar divide, but both are pessimistic about the prospects for dialogue, bridge building, or theoretical synthesis. The picture we paint of IPE in the United States is one of a community distinct from both American IR and British style IPE. The picture is not complete, and we still have more work to do in order to have comparable measures for the UK or Europe. In future research, we could include analysis of data we have collected on Canadian IPE. As the home of Robert Cox and other IPE scholars whose work may more aptly be described as non-American IPE, Canada may offer the bridge that Cohen wants us to walk across. Alternatively, Canadian IR/IPE may simply be divided into European and American camps with equally large gulfs between their respective communities. In other work (Lipson et al., 2007), we suggest that Canada sits in the scholarly middle ground between American and European traditions. Because we only are able to depict the American shore, our data do not allow us to make strong inferences on the probability of overcoming the transatlantic divide in the future. It does allow us to conclude that the picture that Cohen paints of an American school of IPE is largely consistent with our findings. And, given many of the characteristics on which the divide is based, it does seem as though the divide has grown in recent years. When we analyze the responses of IPE scholars to the survey questions we conclude that the consensus in the American school seems increasingly robust on a range of factors that are addressed by Cohen – epistemology, methods, and theory. More strikingly, the analysis of published articles in the top 12 journals provides even stronger evidence that is consistent with Cohen’s story. A complete test would require data on non-American IR/IPE that are similar in quality and form to the data we have presented in this paper. Unfortunately, we currently have very little systematic evidence from surveys or book/journal article analysis on IPE outside the United States. 27
R E V I E W O F I N T E R N AT I O N A L P O L I T I C A L E C O N O M Y
In the summer/fall of 2008 we will field a survey of IR scholars from the UK, Ireland, Canada, Israel, South Africa, New Zealand, Hong Kong, and Singapore. Further we will initiate a journal coding project that will include the next five IR journals that are more closely associated with European IR (beyond EJIR, JPR, and BJPS, which we have already coded). After these data collection efforts we will be able to provide multiple measures for the size of the transatlantic divide and the degree to which it has grown or shrunk over time. We will also be able to ascertain the specific areas in which the divide is large and those where a relatively short bridge would help to realize Cohen’s goal of constructing a more coherent subfield of IPE. ACKNOWLEDGEMENT We would like to thank Kate Weaves for convincing us to write this paper and then corralling an array of diverse scholars to engage the broader questions that we address briefly herein. For comments, inspiration, and tough questions (many of which remain unanswered), we thank Mark Blyth, Phil Cerny, Benjamin Cohen, Marty Finnemore, Judy Goldstein, Robert Keohane, and David Lake. NOTES 1 The lecture was later revised and published in Review of International Political Economy in May 2007. All future citations will refer to Cohen (2007). 2 See, for example, the responses by John Ravenhill and by Richard Higgot and Matthew Watson in the February 2008 issue of Review of International Political Economy. Cohen’s arguments contradicted prior claims that the field of IPE had by the late 1990s more or less reached a consensus on theories, methods, and questions (Frieden and Martin, 2002), while reifying previous assertions of a schism between an American discipline and a British one (Murphy and Nelson, 2001). 3 The original source for the TRIP journal article database was Garand and Giles ranking from 2003. See Garand and Giles (2003). 4 With the data we have at this time, we cannot effectively determine whether or not Cohen’s depiction of the British school is more or less accurate and can offer only limited evidence on the general trends in British or European IPE scholarship (by seeing what European scholars have published in the top IR journals). However, the TRIP project is in the process of preparing a survey of British IR scholars that will be conducted in September and October 2008. A European-wide survey will follow in 2010. Moreover, we plan to expand the journal article database to include the next five leading journals that publish IR articles. These include journals that disproportionately publish work by European scholars and include: Review of International Political Economy, Global Governance, Millennium, Review of International Studies, and Journal of Common Market Studies. 5 For more information on the TRIP project and related databases see Maliniak et al. (2007a, 2007b, 2007c, 2008); Peterson et al. (2005a, 2005b); Lipson et al. (2007); and the TRIP website at http://www.wm.edu/irtheoryandpractice/ trip/.
28
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E
6 For the 2006 survey conducted in the United States we used a list compiled by U.S. News and World Report to identify all four-year colleges and universities in 2005–2006. There were 1,199 such institutions. We also included the Monterey Institute and seven military schools that were not rated by USNWR but have a relatively large number of political science faculty who teach courses on international relations. We then found the IR faculty members teaching at these schools through an extensive series of web searches, email contacts, and phone calls to department chairs, secretaries, and individual scholars. We identified a total of 2,838 individuals who appeared to research and/or teach IR at these institutions. A total of 133 respondents or their agents informed us that they did not belong in the sample because either they had been misidentified and neither taught nor did research in the field of IR, or they had died, changed jobs, or retired. These individuals were not included in our calculation of the response rate. In all, 1,112 scholars responded to the United States version of the survey, either online or through the mail. Certainly, there are additional individuals who were misidentified by our selection process but who never informed us. Hence, our response rate of over 41% is a conservative estimate. 7 For the TRIP journal article database we include data from articles in the 12 leading journals in the field, selected according to their Garand and Giles (2003) ‘impact’ ratings. The journals include American Political Science Review, American Journal of Political Science, British Journal of Political Science, European Journal of International Relations, International Organization, International Security, International Studies Quarterly, Journal of Conflict Resolution, Journal of Peace Research, Journal of Politics, Security Studies, and World Politics. Although Foreign Affairs and Foreign Policy were ranked higher than some of the journals on our list, we did not include them because neither is peer-reviewed. In the IR-specific journals – European Journal of International Relations, International Organization, International Security, International Studies Quarterly, Journal of Conflict Resolution, Journal of Peace Research, Journal of Politics, Security Studies, and World Politics – we code every article in every issue for every year of their publication between 1980 and 2007. In the general political science journals – American Political Science Review, American Journal of Political Science, British Journal of Political Science – we only code those articles that fall within the IR subfield (broadly defined). For this paper we drew a sample of articles for the 26-year time series by coding every article in issue #1 and #3 for every year. This means that our sample is roughly half the size of the entire population of articles. The total number of articles in our sample for this paper was 1806. For more precise rules that were followed see the TRIP codebook that is posted online at: http://mjtier.people.wm.edu/TRIP.pdf 8 Our list of journals may not capture the journals that are perceived as most influential among European scholars. To identify these journals would require a survey. Instead, we rely upon the Garand and Giles (2003) impact ratings, which combine measures of citations per article, familiarity, and strength of evaluation. Murphy and Nelson argue that the two IPEs – British and American – are centered on the two journals: International Organization for the American school, and Review of International Political Economy for the British. Since the TRIP journal article database regrettably does not cover RIPE (it was number 76 on the Garand and Giles rankings in 2003, but by 2007 was ranked 33) this significantly constrains our comparative analysis. However, our data do contain all the articles from the British Journal of Political Science, Journal of Peace Research, and the European Journal of International Relations, which tend to publish proportionately more work from scholars at non-American institutions.
29
R E V I E W O F I N T E R N AT I O N A L P O L I T I C A L E C O N O M Y
9 This number represents those who answered that their primary or secondary area of study/substantive focus was IPE. In 2004 36% of respondents reported IPE as either a primary or secondary area of study. We are surprised and have no explanation for the 6 point drop from 2004 to 2006. 10 The finding on University of Colorado could be called the ‘Leblang effect’, since David Leblang has published so many IPE articles in the leading journals over the past few years. When we analyze only articles published after 2000, we find that Pennsylvania State University and Stanford are the top producers. Also making it into the top ten are LSE (fifth), University of Pennsylvania (fifth), University of California, San Diego (ninth), and University of Illinois, Urbana–Champaign (ninth). Note: there is a five-way tie for ninth. 11 For a detailed analysis of women in the IR discipline and what issue areas they tend to study see Maliniak et al. (2008). 12 Further, many of the women who get named on these three lists have made substantial contributions to research in IPE (Beth Simmons, Helen Milner, Martha Finnemore, and Susan Strange). For the complete lists see Maliniak et al. (2007c). 13 Question reads, ‘List the four journals you read most regularly or otherwise rely on for the best research in your area of expertise’. 14 All citation data come from the Social Science Citation Index. Due to access problems, we collected data only for articles published in or after 1990. The values were recorded in August 2007. 15 For detailed definitions and coding rules for these paradigms see TRIP codebook variable #10. The codebook can be found here: http://mjtier. people.wm.edu/TRIP.pdf. 16 Based on what we know about other IR journals in the field, we expect that if we expanded our sample of journals to include the next 12 journals on the Garand and Giles list we might get different results. Similarly, if we coded books as part of the IPE literature, this would also likely change our findings. However, the labor intensive nature of the coding project, limited time and funds has thus far precluded such analyses. 17 While few commentators would expect realism to dominate the study of IPE, most do assert that it dominates the broader discipline of IR (Doyle, 1997; Vasquez, 1998). While this may have been true at some point in the pre-1980 era, it has never been true since then. For startling evidence of realism’s relatively small proportion of the IR literature see Maliniak et al. (2007a). 18 This variable captures which paradigms are discussed in a serious way – that is, treated as alternative explanations, used to derive testable hypotheses or used to frame the research question. A simple ‘straw-man’ depiction of an alternative paradigm does not qualify as ‘taken seriously’. Instead, the reader needs to learn something about the utility, internal logic, or scope conditions of the alternative paradigm (or a specific model following from some alternative paradigm), in order to be categorized as ‘taken seriously’. The fact that a particular model or theory has implications for a given paradigm does not mean that the article takes that paradigm seriously. With one exception, we DO NOT allow the same value to be entered for ‘paradigms taken seriously by author’ as for ‘paradigm advanced/advocated by author or used to guide analysis’. For example, if an author is advancing a ‘defensive realist’ approach and he/she tests an alternative ‘offensive realist’ approach, then the coder would enter ‘realist’ for ‘paradigm advanced/advocated by author or used to guide analysis’ but not for ‘paradigms taken seriously by author’. 19 This finding is even more striking because there are proportionally fewer realist articles in IPE, and ‘paradigm taken seriously’ can only take on a value other than the paradigm of the article. Therefore, since only 3% of the IPE literature is
30
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E
20
21 22
23 24
realist, 97% of the articles are eligible to have realism as an alternative paradigm taken seriously. Whereas liberal articles represent 66% of the IPE literature, and only 34% of the remaining IPE literature can include liberalism as an alternative paradigm taken seriously. In short, Lipson (1984) may be right, in practice scholars employ liberal theories to explain IPE and they employ realist theories to explain security affairs. Wendt (1987), Dessler (1989), and Waltz (1979) are all examples of analytical/non-formal conceptual articles. We do not code an article this way if it employs any of the empirical methods (Quantitative, Qualitative, Formal Modeling, Counterfactual, Descriptive, Policy Analysis, or Experimental methods). This means that articles with a significant non-formal theoretical component DO NOT get coded as ‘Analytic/Non-formal’ even if they make a significant theoretical contribution (for example, Lake, 2006). Due to the use of rolling averages, the values of zero in 1992 and 1993 imply that 1991 and 1994 also were years in which no Analytic/non-formal articles were published. For interesting and striking recent exceptions see a variety of books that articulate the importance of ideational factors in explaining IPE outcomes (Abdelal, 2001, 2007; Abdelal et al., forthcoming; Best, 2005; Blythe, 2002; Germain, 1997; Hall, 1999; McNamara, 1998; Parsons, 2003; Rupert, 2000; Seabrooke, 2006; Sharman, 2006; Sinclair, 2005; Weaver, 2008). The fact that such research is appearing in books rather than the top 12 journals may result from editorial board screening by an entrenched American IPE mainstream. Alternatively, scholars with ideational arguments may not submit to these journals because they have either stopped reading them or believe their work won’t get a fair hearing. Alternatively, this pattern may follow from the nature of ideational work where book length manuscripts are a more effective format for the kind of qualitative work that is often associated with ideational arguments. Of course, this last possibility would not explain the large increase in ideational arguments published in these same journals outside the IPE subfield. Finally, it may simply be that the 12 journals we coded do not include existing and new journals that tend to attract qualitative work (foremost Review of International Political Economy and New Political Economy). The creation of RIPE in the mid-1990s may have reduced the number of ideational submissions to these other journals. This does not necessarily imply that they are incorporating these into their work in IPE, since they very well may only include these variables in other (non-IPE) projects. Susan Strange (1995: 290) famously argued that IPE was not only dominated by the American school, but that Americans ‘are deaf and blind to anything that’s not published in the USA’.
N OT ES O N C O N T R I B U T O R S Daniel Maliniak is a doctoral student in Political Science at the University of California–San Diego. His interests include international political economy, international relations theory and international organizations. His publications have appeared in Foreign Policy, International Journal, Politics and Gender and Evaluation Review. He also served as the Editorial Administrator for the journal Security Studies from August 2005 to August 2008. Michael J. Tierney is Director of International Relations and Associate Professor of Government at the College of William & Mary. He is co-author or co-editor of two books: Delegation and Agency in International Organizations (Cambridge University
31
R E V I E W O F I N T E R N AT I O N A L P O L I T I C A L E C O N O M Y
Press, 2006), and Greening Aid? Understanding the Environmental Impact of Development Assistance. (Oxford University Press, 2008). He has published articles in International Organization, Foreign Policy, Environment, Law and Contemporary Problems, Politics and Gender, Journal of International Relations and Development, and the International Journal.
REFERENCES Abdelal, R. (2001) National Purpose in the World Economy: Post-Soviet States in Comparative Perspective, Ithaca, NY: Cornell University Press. Abdelal, R. (2007) Capital Rules: The Construction of Global Finance, Cambridge, MA: Harvard University Press. Abdelal, R., Blyth, M. and Parsons, C. (forthcoming) Constructivist Political Economy, Ithaca, NY: Cornell University Press. Best, J. (2005) The Limits of Transparency: Ambiguity and the History of International Finance, Ithaca, NY: Cornell University Press. Blyth, M. (2002) Great Transformation: Economic Ideas and Institutional Change in the Twentieth Century, Cambridge: Cambridge University Press. Checkel, J. (1998) ‘The Constructivist Turn in International Relations Theory’, World Politics, 50: 324–48. Cohen, B. (2007) ‘The Transatlantic Divide: Why are American and British IPE so Different?’, Review of International Political Economy, 14: 197–219. Cohen, B. (2008) International Political Economy: An Intellectual History, Princeton, NJ: Princeton University Press. Dessler, D. (1989) ‘What’s at Stake in the Agent-Structure Debate?’, International Organization, 4: 441–73. Doyle, M. (1997) Ways of War and Peace, New York: Norton. Frieden, J. and Lake, D. (1999) International Political Economy: Perspectives on Global Power and Wealth, London: Routledge. Frieden, J. and Martin, L. (2002) ‘International Political Economy: The State of the Sub-discipline’, in I. Katznelson and H. Milner (eds) Political Science: The State of the Discipline, New York: Norton, pp. 118–46. Garand, J. and Giles, M. (2003) ‘Journals in the Discipline: A Report on a New Survey of American Political Scientists’, PS: Political Science and Politics, April: 293–308. Germain, R. (1997) International Organization of Credit: States and Global Finance in the World Economy, New York: Cambridge University Press. Gilpin, R. (1987) The Political Economy of International Relations, Princeton, NJ: Princeton University Press. Hall, R. B. (1999) National Collective Identity: Social Constructs and International Systems, New York: Columbia University Press. Higgot, R. and Watson, M. (2008) ‘All at Sea in a Barbed Wire Canoe: Professor Cohen’s Transatlantic Voyage in IPE’, Review of International Political Economy, 15: 1–17. Hoffmann, S. (1977) ‘An American Social Science: International Relations’, Daedalus, 106: 41–60. Katzenstein, P., Keohane, R. and Krasner, S. (1998) ‘International Organization and the Study of World Politics’, International Organization, 52: 645–85. Lake, D. (2006) ‘International Political Economy: An Emerging Interdiscipline’, in B. Weingast and D. Wittman (eds)The Oxford Handbook of Political Economy, New York: Oxford University Press. Lipson, C. (1984) ‘International Cooperation in Economic and Security Affairs’, World Politics, 37: 1–23.
32
M A L I N I A K A N D T I E R N E Y: T H E A M E R I C A N S C H O O L O F I P E
Lipson, M., Maliniak, D., Oakes, A., Peterson, S. and Tierney, M. (2007) ‘Divided discipline? Comparing Views of U.S. and Canadian IR Scholars’, International Journal, 62: 327–43. Maliniak, D., Oakes, A., Peterson, S. and Tierney, M. (2007a) ‘The International Relations Discipline: 1980–2007’, Paper presented at the American Political Science Association, Chicago, IL. Maliniak, D., Oakes, A., Peterson, S. and Tierney, M. (2007b) ‘Inside the Ivory Tower’, Foreign Policy, March/April: 62–8. Maliniak, D., Oakes, A., Peterson, S. and Tierney, M. (2007c) ‘The View from the Ivory Tower: TRIP Survey of IR Faculty in the U.S. and Canada’, Institute for the Theory and Practice of International Relations, College of William and Mary. Williamsburg, VA. http://www.wm.edu/ irtheoryandpractice/trip/surveyreport06-07.pdf. Maliniak, D., Oakes, A., Peterson, S. and Tierney, M. (2008) ‘Women in International Relations’, Politics & Gender, 4: 12–144. McNamara, K. (1998) The Currency of Ideas: Monetary Politics in the European Union Ithaca, NY: Cornell University Press. Murphy, C. and Nelson, D. (2001) ‘International Political Economy: A Tale of Two Heterodoxies’, British Journal of Politics and International Relations, 3: 393–412. Parsons, C. (2003) A Certain Idea of Europe, Ithaca, NY: Cornell University Press, Political Economy Series. Peterson, S. and Tierney, M. with Maliniak, D. (2005a) ‘Teaching and Research Practices, Views on the Discipline, and Policy Attitudes of International Relations Faculty at U.S. Colleges and Universities’, Reves Center for International Studies, College of William and Mary, Williamsburg, VA. Peterson, S., Tierney, M. and Maliniak, D. (2005b) ‘Inside the Ivory Tower’, Foreign Policy, November/December: 58–64. Ravenhill, J. (2008) ‘In Search of the Missing Middle’, Review of International Political Economy, 15: 18–29. Rupert, M. (2000) Ideologies of Globalization: Contending Visions of a New World Order, London: Routledge. Seabrooke, L. (2006) The Social Sources of Financial Power: Domestic Legitimacy and International Financial Orders, Ithaca, NY: Cornell University Press. Sharman, J. (2006) Havens in a Storm: The Struggle for Global Tax Regulation, Ithaca, NY: Cornell University Press. Sinclair, T. (2005) The New Masters of Capitalism: American Bond Rating Agencies and the Politics of Creditworthiness, Ithaca, NY: Cornell University Press. Strange, S. (1995) ‘ISA as a Microcosm’, International Studies Quarterly, 39: 289–96. Vasquez, J. (1998) The Power of Power Politics: From Classical Realism to Neotraditionalism, Cambridge: Cambridge University Press. Waever, O. (1998) ‘The Sociology of a Not So International Discipline: American and European Developments in International Relations’, International Organization, 52: 687–727. Waltz, K. (1979) Theory of International Politics, New York: McGraw-Hill. Weaver, C. (2008) Hypocrisy Trap: The World Bank and the Poverty of Reform, Princeton, NJ: Princeton University Press. Weingast, B. and Wittman, D. A. (eds) (2006) The Oxford Handbook of Political Economy, Oxford: Oxford University Press. Wendt, A. (1987) ‘The Agent-Structure Problem in International Relations Theory’, International Organization, 41: 335–70.
33
